sunholo 0.134.4__py3-none-any.whl → 0.134.6__py3-none-any.whl

sunholo/discovery_engine/cli.py

@@ -191,7 +191,7 @@ def import_document_with_metadata_command(args):
 
 def search_command(args):
     """Handles the `discovery-engine search` subcommand (Data Store Chunks)."""
-    console.print(f"[cyan]Searching data store '{args.data_store_id}' for query: '{args.query}' (mode: chunks)[/cyan]")
+    console.print(f"[cyan]Searching data store '{args.data_store_id}' for query: '{args.query}' (mode: {args.content_search_spec_type})[/cyan]")
     try:
         client = DiscoveryEngineClient(
             project_id=args.project,
@@ -208,7 +208,7 @@ def search_command(args):
                 page_size=args.page_size,
                 parse_chunks_to_string=args.parse_chunks_to_string,
                 serving_config=args.serving_config,
-                # data_store_ids=args.data_store_ids # Ensure these args are added to parser if needed
+                filter_str=args.filter,
             )
         elif args.content_search_spec_type == "documents":
             results_data = client.get_documents(
@@ -216,7 +216,7 @@ def search_command(args):
                 page_size=args.page_size,
                 parse_documents_to_string=args.parse_chunks_to_string,
                 serving_config=args.serving_config,
-                # data_store_ids=args.data_store_ids # Ensure these args are added to parser if needed
+                filter_str=args.filter,
             )  
         else:
             raise ValueError("Invalid content_search_spec_type. Must be 'chunks' or 'documents'.")          
@@ -224,34 +224,56 @@ def search_command(args):
         if args.parse_chunks_to_string:
             console.print("\n[bold magenta]--- Combined Chunk String ---[/bold magenta]")
             console.print(results_data if results_data else "[yellow]No results found or error occurred.[/yellow]")
+        elif isinstance(results_data, str):
+            # Handle string result when parse_chunks_to_string is False but a string was returned anyway
+            console.print("\n[bold magenta]--- Results String ---[/bold magenta]")
+            console.print(results_data)
         elif results_data: # It's a pager object
-            console.print("\n[bold magenta]--- Individual Chunks ---[/bold magenta]")
-            chunk_count = 0
-            try:
-                 # Iterate through the pager returned by get_chunks
-                 for page in results_data.pages:
-                     if not hasattr(page, 'results') or not page.results: continue
-                     for result in page.results:
-                          # Ensure the result structure is as expected by get_chunks
-                          if hasattr(result, 'chunk'):
-                               chunk_count += 1
-                               console.print(f"\n[bold]Chunk {chunk_count}:[/bold]")
-                               # Use the client's formatter if available
-                               console.print(client.chunk_format(result.chunk))
-                          elif hasattr(result, 'document') and hasattr(result.document, 'chunks'):
-                               # Fallback if structure is different (e.g., document with chunks)
-                               for chunk in result.document.chunks:
-                                    chunk_count += 1
-                                    console.print(f"\n[bold]Chunk {chunk_count} (from doc {result.document.id}):[/bold]")
-                                    console.print(f"  Content: {getattr(chunk, 'content', 'N/A')}")
-                                    console.print(f"  Doc Name: {getattr(chunk, 'document_metadata', {}).get('name', 'N/A')}") # Example access
-
-                 if chunk_count == 0:
-                     console.print("[yellow]No chunks found in the results.[/yellow]")
-
-            except Exception as page_err:
-                 console.print(f"[bold red]Error processing search results pager: {page_err}[/bold red]")
-                 console.print(f"[red]{traceback.format_exc()}[/red]")
+            if args.content_search_spec_type == "chunks":
+                console.print("\n[bold magenta]--- Individual Chunks ---[/bold magenta]")
+                chunk_count = 0
+                try:
+                    # Iterate through the pager returned by get_chunks
+                    for page in results_data.pages:
+                        if not hasattr(page, 'results') or not page.results: continue
+                        for result in page.results:
+                            # Ensure the result structure is as expected by get_chunks
+                            if hasattr(result, 'chunk'):
+                                chunk_count += 1
+                                console.print(f"\n[bold]Chunk {chunk_count}:[/bold]")
+                                # Use the client's formatter if available
+                                console.print(client.chunk_format(result.chunk))
+                            elif hasattr(result, 'document') and hasattr(result.document, 'chunks'):
+                                # Fallback if structure is different (e.g., document with chunks)
+                                for chunk in result.document.chunks:
+                                        chunk_count += 1
+                                        console.print(f"\n[bold]Chunk {chunk_count} (from doc {result.document.id}):[/bold]")
+                                        console.print(f"  Content: {getattr(chunk, 'content', 'N/A')}")
+                                        console.print(f"  Doc Name: {getattr(chunk, 'document_metadata', {}).get('name', 'N/A')}") 
+                    if chunk_count == 0:
+                        console.print("[yellow]No chunks found in the results.[/yellow]")
+
+                except Exception as page_err:
+                    console.print(f"[bold red]Error processing search results pager: {page_err}[/bold red]")
+                    console.print(f"[red]{traceback.format_exc()}[/red]")
+            elif args.content_search_spec_type == "documents":
+                console.print("\n[bold magenta]--- Individual Documents ---[/bold magenta]")
+                doc_count = 0
+                try:
+                    # Iterate through the pager returned by get_documents
+                    for page in results_data.pages:
+                        if not hasattr(page, 'results') or not page.results: continue
+                        for result in page.results:
+                            if hasattr(result, 'document'):
+                                doc_count += 1
+                                console.print(f"\n[bold]Document {doc_count}:[/bold]")
+                                console.print(client.document_format(result.document))
+                    
+                    if doc_count == 0:
+                        console.print("[yellow]No documents found in the results.[/yellow]")
+                except Exception as page_err:
+                    console.print(f"[bold red]Error processing document results: {page_err}[/bold red]")
+                    console.print(f"[red]{traceback.format_exc()}[/red]")
         else:
             console.print("[yellow]No results found or error occurred.[/yellow]")
 
@@ -519,14 +541,17 @@ def setup_discovery_engine_subparser(subparsers):
     import_doc_meta_parser.add_argument('--branch', default='default_branch', help='Target branch')
     import_doc_meta_parser.set_defaults(func=discovery_engine_command)
 
-    # --- Search Data Store (Chunks) subcommand ---
-    search_parser = discovery_engine_subparsers.add_parser('search', help='Search a datastore (fetches chunks)')
+    # --- Search Data Store (Chunks/Documents) subcommand ---
+    search_parser = discovery_engine_subparsers.add_parser('search', help='Search a datastore (fetches chunks or documents)')
     search_parser.add_argument('--query', required=True, help='The search query')
     search_parser.add_argument('--data-store-id', required=True, help='Data store ID to search')
     search_parser.add_argument('--page-size', type=int, default=10, help='Max results per page')
     search_parser.add_argument('--parse-chunks-to-string', action='store_true', help='Output results as one formatted string. Only applicable for "chunks"')
     search_parser.add_argument('--serving-config', default='default_config', help='Serving config ID for the data store')
     search_parser.add_argument('--content_search_spec_type', default="chunks", help='"chunks" or "documents" depending on data store type')
+    search_parser.add_argument('--filter', help='filter for the search')
+
+
     # Add arguments for num_previous_chunks, num_next_chunks, data_store_ids if needed
     # search_parser.add_argument('--num-previous-chunks', type=int, default=3)
     # search_parser.add_argument('--num-next-chunks', type=int, default=3)

sunholo/discovery_engine/discovery_engine_client.py

@@ -213,6 +213,7 @@ class DiscoveryEngineClient:
         parse_chunks_to_string: bool = True,
         serving_config: str = "default_serving_config",
         data_store_ids: Optional[List[str]] = None,
+        filter_str:str=None
     ):
         """Retrieves chunks or documents based on a query.
 
@@ -237,10 +238,10 @@ class DiscoveryEngineClient:
                     print(f"Chunk: {chunk.snippet}, document name: {chunk.document_name}")
             ```
         """
-        # Use search_with_filters with filter_str=None to perform a regular search
+
         return self.search_with_filters(
             query=query,
-            filter_str=None,
+            filter_str=filter_str,
             num_previous_chunks=num_previous_chunks,
             num_next_chunks=num_next_chunks,
             page_size=page_size,
@@ -259,6 +260,7 @@ class DiscoveryEngineClient:
         parse_chunks_to_string: bool = True,
         serving_config: str = "default_serving_config",
         data_store_ids: Optional[List[str]] = None,
+        filter_str:str=None
     ):
         """Asynchronously retrieves chunks or documents based on a query.
 
@@ -372,6 +374,7 @@ class DiscoveryEngineClient:
         parse_documents_to_string: bool = True,
         serving_config: str = "default_serving_config",
         data_store_ids: Optional[List[str]] = None,
+        filter_str:str=None
     ):
         """Retrieves entire documents based on a query.
 
@@ -397,7 +400,7 @@ class DiscoveryEngineClient:
         # Use search_with_filters with content_search_spec_type="documents" to get documents instead of chunks
         return self.search_with_filters(
             query=query,
-            filter_str=None,
+            filter_str=filter_str,
             page_size=page_size,
             parse_chunks_to_string=parse_documents_to_string,
             serving_config=serving_config,
@@ -412,6 +415,7 @@ class DiscoveryEngineClient:
         parse_documents_to_string: bool = True,
         serving_config: str = "default_serving_config",
         data_store_ids: Optional[List[str]] = None,
+        filter_str:str=None
     ):
         """Asynchronously retrieves entire documents based on a query.
 
@@ -430,7 +434,7 @@ class DiscoveryEngineClient:
         # as it doesn't currently have that parameter
         return await self.async_search_with_filters(
             query=query,
-            filter_str=None,
+            filter_str=filter_str,
             page_size=page_size,
             parse_chunks_to_string=parse_documents_to_string, 
             serving_config=serving_config,
@@ -836,17 +840,22 @@ class DiscoveryEngineClient:
             log.info(f"No results {search_request.data_store_specs=}: {str(e)}")
             return None
         
-        if content_search_spec_type=="chunks":
-            if parse_chunks_to_string:
-                big_string = self.process_chunks(search_response)
-                log.info(f"Discovery engine chunks string sample: {big_string[:100]}")
+        if parse_chunks_to_string:
+            if content_search_spec_type=="chunks":
+                if parse_chunks_to_string:
+                    big_string = self.process_chunks(search_response)
+                    log.info(f"Discovery engine chunks string sample: {big_string[:100]}")
+
+                    return big_string
+                
+            elif content_search_spec_type=="documents":
+                big_string = self.process_documents(search_response)
+                log.info(f"Discovery engine documents string sample: {big_string[:100]}")
+
                 return big_string
-        elif content_search_spec_type=="documents":
-            big_string = self.process_documents(search_response)
-            log.info(f"Discovery engine documents string sample: {big_string[:100]}")
-            return big_string
         
         log.info("Discovery engine response object")
+        
         return search_response
 
     async def async_search_with_filters(self, query, filter_str=None,

{sunholo-0.134.4.dist-info → sunholo-0.134.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sunholo
-Version: 0.134.4
+Version: 0.134.6
 Summary: AI DevOps - a package to help deploy GenAI to the Cloud.
 Author-email: Holosun ApS <multivac@sunholo.com>
 License: Apache License, Version 2.0

{sunholo-0.134.4.dist-info → sunholo-0.134.6.dist-info}/RECORD

@@ -73,9 +73,9 @@ sunholo/database/sql/sb/return_sources.sql,sha256=89KAnxfK8n_qGK9jy1OQT8f9n4uYUt
 sunholo/database/sql/sb/setup.sql,sha256=CvoFvZQev2uWjmFa3aj3m3iuPFzAAJZ0S7Qi3L3-zZI,89
 sunholo/discovery_engine/__init__.py,sha256=hLgqRDJ22Aov9o2QjAEfsVgnL3kMdM-g5p8RJ9OyKdQ,130
 sunholo/discovery_engine/chunker_handler.py,sha256=wkvXl4rFtYfN6AZUKdW9_QD49Whf77BukDbO82UwlAg,7480
-sunholo/discovery_engine/cli.py,sha256=bUhCPoKrkMtdeTBHLyCZf9syVwHn5kE0yXpqDBIzmTc,34120
+sunholo/discovery_engine/cli.py,sha256=tsKqNSDCEsDTz5-wuNwjttb3Xt35D97-KyyEiaqolMQ,35628
 sunholo/discovery_engine/create_new.py,sha256=WUi4_xh_dFaGX3xA9jkNKZhaR6LCELjMPeRb0hyj4FU,1226
-sunholo/discovery_engine/discovery_engine_client.py,sha256=Ak3VpadtgpPWfIEot87EiNh4vbDUg9gQVa-1UDnoGMA,58442
+sunholo/discovery_engine/discovery_engine_client.py,sha256=XOKPx2C9sXSx1Z1_23IynrA19NTzWwcXFdvZm-jYXQY,58589
 sunholo/discovery_engine/get_ai_search_chunks.py,sha256=I6Dt1CznqEvE7XIZ2PkLqopmjpO96iVEWJJqL5cJjOU,5554
 sunholo/embedder/__init__.py,sha256=sI4N_CqgEVcrMDxXgxKp1FsfsB4FpjoXgPGkl4N_u4I,44
 sunholo/embedder/embed_chunk.py,sha256=did2pKkWM2o0KkRcb0H9l2x_WjCq6OyuHDxGbITFKPM,6530
@@ -168,9 +168,9 @@ sunholo/vertex/init.py,sha256=1OQwcPBKZYBTDPdyU7IM4X4OmiXLdsNV30C-fee2scQ,2875
 sunholo/vertex/memory_tools.py,sha256=tBZxqVZ4InTmdBvLlOYwoSEWu4-kGquc-gxDwZCC4FA,7667
 sunholo/vertex/safety.py,sha256=S9PgQT1O_BQAkcqauWncRJaydiP8Q_Jzmu9gxYfy1VA,2482
 sunholo/vertex/type_dict_to_json.py,sha256=uTzL4o9tJRao4u-gJOFcACgWGkBOtqACmb6ihvCErL8,4694
-sunholo-0.134.4.dist-info/licenses/LICENSE.txt,sha256=SdE3QjnD3GEmqqg9EX3TM9f7WmtOzqS1KJve8rhbYmU,11345
-sunholo-0.134.4.dist-info/METADATA,sha256=g2a99cYiyPfDeVa55srGe2k_5_MHsJhzwBll43hQ56M,10067
-sunholo-0.134.4.dist-info/WHEEL,sha256=pxyMxgL8-pra_rKaQ4drOZAegBVuX-G_4nRHjjgWbmo,91
-sunholo-0.134.4.dist-info/entry_points.txt,sha256=bZuN5AIHingMPt4Ro1b_T-FnQvZ3teBes-3OyO0asl4,49
-sunholo-0.134.4.dist-info/top_level.txt,sha256=wt5tadn5--5JrZsjJz2LceoUvcrIvxjHJe-RxuudxAk,8
-sunholo-0.134.4.dist-info/RECORD,,
+sunholo-0.134.6.dist-info/licenses/LICENSE.txt,sha256=SdE3QjnD3GEmqqg9EX3TM9f7WmtOzqS1KJve8rhbYmU,11345
+sunholo-0.134.6.dist-info/METADATA,sha256=olUgsXM6SFdKO9eS9oAh8I5EZ1zLhm8i_lEJ4WxoZak,10067
+sunholo-0.134.6.dist-info/WHEEL,sha256=pxyMxgL8-pra_rKaQ4drOZAegBVuX-G_4nRHjjgWbmo,91
+sunholo-0.134.6.dist-info/entry_points.txt,sha256=bZuN5AIHingMPt4Ro1b_T-FnQvZ3teBes-3OyO0asl4,49
+sunholo-0.134.6.dist-info/top_level.txt,sha256=wt5tadn5--5JrZsjJz2LceoUvcrIvxjHJe-RxuudxAk,8
+sunholo-0.134.6.dist-info/RECORD,,